Talk:Reading Town Hall

Close paraphrasing
This article seems to contain content too closely paraphrased from this pdf. While facts are not copyrightable, creative elements of presentation - including both structure and language - are. For an example of close paraphrasing, consider the following:


 * The source says:

There are other passages that similarly follow too closely. I note especially additional issues in the second to last paragraph, which begins "In 1943...an air raid destroyed." If we bold the content from that paragraph that is precisely duplicated from that external source, it looks like this:

So that it will not constitute a derivative work, this article should be rewritten in the temporary space that is now linked from the article's front. The essay Close paraphrasing contains some suggestions for rewriting that may help avoid these issues. The article Wikipedia Signpost/2009-04-13/Dispatches, while about plagiarism rather than copyright concerns, also contains some suggestions for reusing material from sources that may be helpful, beginning under "Avoiding plagiarism".

Alternatively, if the material can be verified to be public domain or permission is provided, we can use the original text with proper attribution.

Please let me know if you have questions about this. --Moonriddengirl (talk) 17:20, 17 September 2010 (UTC)


 * I have to say that I don't think my text was that closely paraphrased from my source, although there may have been some inadvertent mental reuse of text. I note that one of your examples above neatly uses an ellipsis to maximise the apparent similarity, which I cannot help thinking is a little dubious.


 * Nonetheless I took the advice of one of the essays you reference above to internalise several sources before rewriting (in the temp area). If nothing else, this has improved the articles coverage and citations. But in some cases there are not that many ways of expressing a simple fact in a few words, which does make it quite hard to avoid the appearance of plagiarism at the level of test you are applying. Despite the fact that the sources I used are quite disparate, with different authors and different eras (the listing texts date back to the 50s, the history book was published in 1980, and the PDF you refer to is presumably fairly recent), there are clear cases where they use exactly the same sentence structures and (almost) the same words.


 * Anyway I think what I've done is the best I can manage, so take a look and see if passes. By the way, I only rewrote the history section, as that is where you applied the tag. The article lead is unchanged. -- Starbois (talk) 20:52, 19 September 2010 (UTC)


 * Err, is anybody going to do anything about this. The article has been blanked out with an ugly looking template for eight days now. In any case there has been replacement text in the temporary area for six days, and AFAICT nobody has even read it yet. Certainly nobody has responded to my request for comment above. If "an administrator or an OTRS agent" doesn't do something soon, I shall find it difficult to resist the temptation to be bold. -- Starbois (talk) 16:42, 25 September 2010 (UTC)
 * These items are listed for eight days before administrator closure to allow time for material to be rewritten or for permission to process. As it happens, this one has come due for closure today. It is standard procedure for the template to remain on the article during that time. I'm sorry that I did not respond to your earlier note. I should obviously clarify the "form letter" I used above to specify that questions should be left at my talk page. I typically process dozens of copyright concerns in a day, and I do not watchlist most of the articles in question. I'm afraid I had not noticed your response until the listing came current.


 * Before I process this, I'll offer first to invite another copyright administrator to take a look at the original, if you would like a third opinion as to whether content is similar enough to the source to constitute a copyright concern. (It occurs to me that you may not have known that mine is a second opinion, since you may not have known about the DYK nomination of this article; see the listing.)


 * I will explain, though, that the addition or subtraction of material does not efface copyright concerns, which is why I did indeed edit it to make the similarity clear. Under the United States law that governs Wikipedia, a "derivative work" is a work that transforms another, already copyrighted work. This transformation includes, but is not limited to, abridgment, editorial revisions, annotations, and elaborations. You cannot prepare a derivative work without permission of the copyright holder of the original material. If you make substantial use of a copyrighted work, you stand at risk of copyright infringement, and accordingly our copyright policy requires that all content you import from non-free sources must be written from scratch, except as you incorporate brief, clearly marked quotations.


 * Anyway, just let me know if you'd like me to ask another admin to close this one. If not, I'll take a look at the temp page and use it to replace the problematic content if it's fully rewritten. (And I am watching now. :)) --Moonriddengirl (talk) 22:23, 25 September 2010 (UTC)


 * Thank you for your response. I'm happy for you to process it. I'm reasonably confident that there is nothing in the new version that can be seen as a derivative work, although in truth I would probably have said the same about the original before your post.


 * I'm interested in your comments on a derived work. Obviously, given its aversion to original research, everything in WP is at least in some sense derived. And as, in real life, a software developer, I am necessarily fairly familiar with IPR and copyright issues. I guess the structured syntax of a computer language makes detecting derivation a more clean cut distinction than in natural languages though. Still real life calls, so I'd better not debate it further here and now.-- Starbois (talk) 10:45, 27 September 2010 (UTC)

Thank you for the rewriting. I don't see any problems with it and have (as you'll no doubt notice) put it in place.

Balancing Wikipedia's requirement that we rewrite from scratch with Wikipedia's requirement that we not deviate from our sources causes headaches for quite a few people, and your question is raised so frequently that I kind of wonder if I should put something in our copyright FAQ about it. :) It's a good question. It's just not that easy to answer. Frequently, it can be at least partially addressed by the use of brief quotations from our sources when we can come up with good reasons for them. Use of quotations must be transformative, so we have to avoid the impulse to use something just because they said it well. Very occasionally I will indulge in an "as John Smith described it" quote, but most of the time our quotes should (as per WP:NFC) "illustrate a point, establish context, or attribute a point of view or idea."

When looking out for close paraphrasing, I personally look for a combination of "fragmented literal similarity" and "comprehensive non-literal similarity": if we see stretches of similar language in a work that overall follows the same structure of the original, the risk of actionable "close paraphrasing" rises. You balance that against the creativity of either; structure is uncreative if anybody would have put the same details in the same order. Language is uncreative if it is formulaic. "John Smith was born on 27 September, 2010, in Paris, France" is uncreative. The more language and details you stack, though, the more likely the content is to become creative. The threshold for creativity in the U.S. (which governs us) is pretty low. When writing, I work to vary not only my language but also the organization of material. U.S. courts have found copyright violation even when none of the language of the original is retained, if the "structure and pattern" are similar. And, yes, it sometimes feels like reaching behind your back to touch your elbow. :/ Keeping the information true to our sources without following them can be a challenge. The more sources we find the better, since we're far less likely to accidentally appropriate the structure of a source if we're weaving together facts from multiple sources.

Happy to discuss this further, here or at a user talk page if you like, and if you feel the conversation is done, we can collapse or archive this, since it is no longer actively needed. :) --Moonriddengirl (talk) 12:13, 27 September 2010 (UTC)